AI Course Generator

AI-powered educational content creation tool using QuantaLogic Flow workflows and multi-provider AI integration. Generates structured courses in PDF, DOCX, EPUB, and Markdown formats with customizable difficulty levels and comprehensive MCQ support.

Key Entry Points

Core Files:

• ai_course_generator/generate_course.py - CLI interface with Typer framework
• ai_course_generator/course_generator_agent.py - Main workflow engine and AI integration
• pyproject.toml - Build configuration and dependencies

Quick Build & Run Commands

# Install dependencies (recommended)
poetry install

# Generate a course (example)
poetry run python -m ai_course_generator.generate_course --subject "Python" --level beginner --number-of-chapters 5 --words-by-chapter 500

# Alternative installation
pip install -r requirements.txt

CLI Usage

The main entry point is a Typer-based CLI:

poetry run python -m ai_course_generator.generate_course --help

Options include:

• --subject / -s: Course subject
• --number-of-chapters / -n: Number of chapters
• --level / -l: Difficulty level (beginner/intermediate/advanced)
• --words-by-chapter / -w: Target word count per chapter
• --target-directory / -o: Output directory
• --pdf/--no-pdf: Generate PDF output
• --docx/--no-docx: Generate DOCX output
• --epub/--no-epub: Generate EPUB output
• --model-name / -m: AI model to use
• --interactive / -i: Interactive mode

Workflow Overview

flowchart TD
    %% High contrast, colorblind-friendly palette
    classDef start fill:#2c3e50,stroke:#34495e,stroke-width:3px,color:#ffffff;
    classDef process fill:#3498db,stroke:#2980b9,stroke-width:2px,color:#ffffff;
    classDef decision fill:#e74c3c,stroke:#c0392b,stroke-width:2px,color:#ffffff;
    classDef complete fill:#27ae60,stroke:#229954,stroke-width:3px,color:#ffffff;

    S([Start Course Generation]):::start
    V[Validate Input Parameters]:::process
    T[Generate Course Title]:::process
    O[Generate Course Outline]:::process
    L[Generate Chapter Content]:::process
    M{More Chapters to Process?}:::decision
    C[Compile Full Course]:::process
    F[Convert to Output Formats\nPDF, DOCX, EPUB]:::process
    E([Complete]):::complete

    S --> V
    V --> T
    T --> O
    O --> L
    L --> M
    M -- Yes --> L
    M -- No --> C
    C --> F
    F --> E

Loading

Sequence of Operations

sequenceDiagram
    participant U as 👤 User
    participant CLI as 🖥️ CLI Interface
    participant WF as ⚙️ QuantaLogic Workflow
    participant AI as 🤖 AI Model (LLM)
    participant FS as 📁 File System

    U->>CLI: Submit course parameters
    CLI->>WF: Initialize workflow
    
    WF->>AI: Generate course title
    AI-->>WF: Return title
    
    WF->>AI: Generate course outline
    AI-->>WF: Return structured outline
    
    rect rgb(240, 248, 255)
        Note over WF,AI: Chapter Generation Loop
        loop For each chapter
            WF->>AI: Generate chapter content
            AI-->>WF: Return chapter text
            WF->>FS: Save chapter file
        end
    end
    
    WF->>WF: Compile full course
    WF->>FS: Generate output formats
    FS-->>CLI: Course files ready
    CLI-->>U: Course generation complete

Loading

Dependencies

• Uses quantalogic-flow >=0.6.3 (replaces deprecated quantalogic)
• See pyproject.toml and requirements.txt for full list

Documentation

LLM-Optimized Technical Reference:

• docs/project-overview.md - Project purpose, technology stack, and platform requirements
• docs/architecture.md - System organization, workflow components, and data flow with specific file references
• docs/build-system.md - Build instructions, dependencies, and platform setup for all environments
• docs/development.md - Code patterns, development workflows, and implementation examples from codebase
• docs/testing.md - Testing approach, commands, and CI/CD integration
• docs/deployment.md - Packaging, distribution, and production deployment strategies
• docs/files.md - Comprehensive file catalog with descriptions and dependency relationships

---

For advanced technical details, see TECHNICAL_README.md.